The goal of this proposal is to investigate the relationship between skin pigmentation and Parkinson's disease (PD). Early clinical and epidemiologic observations suggested that PD might be most common in persons of European origin. A biologically plausible explanation for this observation emerged in 1983 when the Parkinsonism-causing toxicant MPTP was identified. The toxic form of this compound, MPP+, was shown to bind to neural melanin; leading to the proposal that dermal melanin may "protectively" bind potential neurotoxicants and prevent their entry into the brain. This possibility is further supported by the observation that dermal melanin selectively binds reactive species (free radicals). Preliminary data suggests that persons with greater tendency to sunburn and persons with lighter skin tone as measured by calorimetry are at greater risk of PD. We propose to confirm (or refute) this observation by investigating the relationship between skin color (which directly correlates with dermal melanin concentration) and risk of PD in a unique human population of farmers in the Agricultural Health Study (AHS). The AHS includes approximately 52,000 pesticide applicators and 32,000 of their spouses, living in Iowa and North Carolina for who detailed life histories of exposures to specific pesticides and direct exposure measurements have been collected. The cohort was enrolled in 1994-96, at which time information was collected on pesticide use, other farming practices, and life-style and health, including Parkinson's disease. FAME is an NIH-funded nested case-control study within the AHS comparing persons with Parkinson's disease to controls is in progress. The current proposal will take advantage of this unique resource to investigate the hypothesis that dermal melanin protects against Parkinson's disease. Skin pigmentation will be measured directly using tristimulus calorimetry during on the farm visits. Standardized and validated questions will be used to assess complexion, hair color, and tendency to sunburn. Lighter complexion and lighter hair color (blond, red) and greater tendency to sunburn are expected to be associated with increased risk of PD. We expect to identify 160 Parkinson's disease cases and 3 controls per case in FAME. Data analysis will utilize univariate and multivariate likelihood-based methods for matched case-control studies. If less skin pigmentation, alone or in combination with other risk factors, is shown to increase PD risk, this will suggest a role for melanin in the pathogenesis of PD. This could be an enormous breakthrough in our understanding of factors that influence the risk (and perhaps even the cause) of Parkinson's disease, and could lead to new strategies for disease prevention.